Abstract:

In the path planning problem of Automated Guided Vehicle（AGV） in static environment, because there are too many nodes in the path planning, these paths reduce the running efficiency of the vehicle and increase the wear of the vehicle. In order to solve these problems, a new grid modeling method is proposed to build the operating environment of vehicle, and the path planning of the new grid environment is carried out by using ant colony algorithm, then the convergence of the ant colony algorithm is proved. The vertex of obstacle is extracted from the grid graph as a new alternative point to plan the path, which reduces the number of nodes to be searched in the ant colony algorithm. The results show that the new grid modeling method can be applied to the path planning of ant colony algorithm, which improves the convergence rate of the algorithm, reduces the number of inflection nodes of AGV, and is better than the traditional grid modeling method in comprehensive performance.

Key words: feature extraction, path planning, ant colony algorithm, grid modeling method

摘要：

在静态环境下的自动导引车辆（Automated Guided Vehicle，AGV）路径规划问题中，由于规划的路径节点过多，导致车辆的运行效率降低，损耗增加。为此，提出了一种新的栅格建模方法来模拟车辆运行环境，利用蚁群算法对新的栅格环境进行路径规划，对蚁群算法的收敛性进行证明。从栅格图中提取出障碍物的顶点作为新的备选点来规划路径，减少了蚁群算法中要搜索的节点数目。结果表明，新的栅格法建模可以应用于蚁群算法的路径规划中，并提高算法的收敛速度，减少AGV方向变化的次数，在综合性能上优于传统栅格建模方法。

关键词: 特征提取, 路径规划, 蚁群算法, 栅格法建模